Batches of wddv rewened to 18% or 24% moisture content were dried in a sealed rotating drum containing ; preheated sand. For each initial moisture content, all combinations of the following variables were replicated three times: sand patilcle size (0.3 and 0.6mm). residence time (60, 120 and 180 s), sand-to-grain mass ration ( 3:1, 6:1 and 9:1) and initial sand temperture ( 100, 150 and 200°). The dried paddy was shelled and milled and the percentage of whole grain kernels was determined. In general. the paddy starting at 18% mc led to lower final moisture contents than that starting at 24%. For equivalent final moisture contents, milling yields of whole rice were equivalent. Best yields were generally associated with lower energy input (ie. low to intermediate initial sand temperature and sand-to-grain mass ratios below 4.7:1) Residence time was not critical in determining milling yieid. This study indicates that particulate medium drying of paddy is an interesting alternative to traditional drying since it greatly accelerates drying timee without reducing milling yields. Solar preheating of the sand could permit substantial energy savings.